Bitcoin Price Prediction
Import Libaries
# All coin history data was sourced from [Crypto Download Data](https://www.cryptodatadownload.com/)
library(ggplot2)
library(dplyr)##
## Attaching package: 'dplyr'## The following objects are masked from 'package:stats':
##
## filter, lag## The following objects are masked from 'package:base':
##
## intersect, setdiff, setequal, unionlibrary(devtools)## Loading required package: usethislibrary(plotly)##
## Attaching package: 'plotly'## The following object is masked from 'package:ggplot2':
##
## last_plot## The following object is masked from 'package:stats':
##
## filter## The following object is masked from 'package:graphics':
##
## layoutlibrary(quantmod)## Loading required package: xts## Loading required package: zoo##
## Attaching package: 'zoo'## The following objects are masked from 'package:base':
##
## as.Date, as.Date.numeric##
## Attaching package: 'xts'## The following objects are masked from 'package:dplyr':
##
## first, last## Loading required package: TTR## Registered S3 method overwritten by 'quantmod':
## method from
## as.zoo.data.frame zoo## Version 0.4-0 included new data defaults. See ?getSymbols.library(xts)
library(coinmarketcapr)
library(tidyverse)## ── Attaching packages ─────────────────────────────────────── tidyverse 1.3.0 ──## ✓ tibble 3.0.4 ✓ purrr 0.3.4
## ✓ tidyr 1.1.2 ✓ stringr 1.4.0
## ✓ readr 1.4.0 ✓ forcats 0.5.0## ── Conflicts ────────────────────────────────────────── tidyverse_conflicts() ──
## x plotly::filter() masks dplyr::filter(), stats::filter()
## x xts::first() masks dplyr::first()
## x dplyr::lag() masks stats::lag()
## x xts::last() masks dplyr::last()library(PerformanceAnalytics)##
## Attaching package: 'PerformanceAnalytics'## The following object is masked from 'package:graphics':
##
## legendEvaluating Bitcoin
Clean & Read Historical Data
# Read CSV data into R
btc_data <- read.csv("Coinbase_BTCUSD_d.csv", header = TRUE)
# Display the first 6 elements to ensure that the data is read
head(btc_data)## Timestamp Date Symbol Open High Low Close Volume.BTC
## 1 1605830400 2020-11-20 BTCUSD 17821.58 18239.00 17764.76 18142.52 3909.44
## 2 1605744000 2020-11-19 BTCUSD 17782.91 18193.29 17356.00 17821.58 17141.49
## 3 1605657600 2020-11-18 BTCUSD 17679.36 18488.00 17205.02 17782.91 32425.64
## 4 1605571200 2020-11-17 BTCUSD 16726.64 17880.00 16575.42 17679.36 25230.04
## 5 1605484800 2020-11-16 BTCUSD 15966.89 16892.00 15879.00 16726.64 13948.06
## 6 1605398400 2020-11-15 BTCUSD 16082.01 16175.60 15796.09 15966.89 6250.08
## Volume.USD
## 1 70437003
## 2 306201498
## 3 579119955
## 4 436549314
## 5 230076772
## 6 99871183# Order rows by date
btc = btc_data[order(btc_data$Date),]
head(btc)## Timestamp Date Symbol Open High Low Close Volume.BTC
## 2170 1417392000 2014-12-01 BTCUSD 300.0 370.0 300.00 370.0 0.05656
## 2169 1417478400 2014-12-02 BTCUSD 370.0 378.0 370.00 378.0 15.01000
## 2168 1417564800 2014-12-03 BTCUSD 378.0 378.0 377.01 378.0 0.54660
## 2167 1417651200 2014-12-04 BTCUSD 378.0 378.0 377.10 377.1 0.01000
## 2166 1417737600 2014-12-05 BTCUSD 377.1 377.1 377.10 377.1 0.00000
## 2165 1417824000 2014-12-06 BTCUSD 377.1 378.0 377.10 378.0 0.01500
## Volume.USD
## 2170 19.53
## 2169 5675.07
## 2168 206.52
## 2167 3.77
## 2166 0.00
## 2165 5.67# Remove Timestamp & Symbol
btc = subset(btc, select = -c(Timestamp, Symbol))
# Check
head(btc)## Date Open High Low Close Volume.BTC Volume.USD
## 2170 2014-12-01 300.0 370.0 300.00 370.0 0.05656 19.53
## 2169 2014-12-02 370.0 378.0 370.00 378.0 15.01000 5675.07
## 2168 2014-12-03 378.0 378.0 377.01 378.0 0.54660 206.52
## 2167 2014-12-04 378.0 378.0 377.10 377.1 0.01000 3.77
## 2166 2014-12-05 377.1 377.1 377.10 377.1 0.00000 0.00
## 2165 2014-12-06 377.1 378.0 377.10 378.0 0.01500 5.67# Convert Date factor into date format
btc$Date <- as.Date(btc$Date, format = "%Y-%m-%d")
head(btc)## Date Open High Low Close Volume.BTC Volume.USD
## 2170 2014-12-01 300.0 370.0 300.00 370.0 0.05656 19.53
## 2169 2014-12-02 370.0 378.0 370.00 378.0 15.01000 5675.07
## 2168 2014-12-03 378.0 378.0 377.01 378.0 0.54660 206.52
## 2167 2014-12-04 378.0 378.0 377.10 377.1 0.01000 3.77
## 2166 2014-12-05 377.1 377.1 377.10 377.1 0.00000 0.00
## 2165 2014-12-06 377.1 378.0 377.10 378.0 0.01500 5.67No irregularities to correct. The substantial variance between prices is due to Bitcoin’s rapid ascent in price over the duration of the dataset. As shown in the next section. ## Historical Price Chart
ds <- data.frame(Date = btc$Date, btc$Close, btc$Volume.USD)
ay <- list(
tickfont = list(color = "Orange"),
overlaying = "y",
side = "right",
title = "Trade Volume"
)
fig <- plot_ly(ds, x = ~Date)
fig <- fig %>% add_lines(y = ~btc$Close, name = "Price")
fig <- fig %>% add_lines(y = ~btc$Volume.USD, name = "Volume", yaxis = "y2")
fig <- fig %>% layout(
title = "Bitcoin Price & Trade Volume",
yaxis2 = ay,
xaxis = list(
rangeselector = list(
buttons = list(
list(
count = 3,
label = "3 mo",
step = "month",
stepmode = "backward"),
list(
count = 6,
label = "6 mo",
step = "month",
stepmode = "backward"),
list(
count = 1,
label = "1 yr",
step = "year",
stepmode = "backward"),
list(
count = 1,
label = "YTD",
step = "year",
stepmode = "todate"),
list(step = "all"))),
rangeslider = list(type = "date")),
yaxis = list(
title = "Price (USD)",
tickfont = list(color = "Blue"))
)
fig## Warning: `arrange_()` is deprecated as of dplyr 0.7.0.
## Please use `arrange()` instead.
## See vignette('programming') for more help
## This warning is displayed once every 8 hours.
## Call `lifecycle::last_warnings()` to see where this warning was generated.summary(btc)## Date Open High Low
## Min. :2014-12-01 Min. : 120.0 Min. : 184.0 Min. : 0.06
## 1st Qu.:2016-05-26 1st Qu.: 468.1 1st Qu.: 482.6 1st Qu.: 459.44
## Median :2017-11-19 Median : 3886.6 Median : 3989.5 Median : 3777.14
## Mean :2017-11-22 Mean : 4770.1 Mean : 4906.5 Mean : 4617.20
## 3rd Qu.:2019-05-27 3rd Qu.: 8282.1 3rd Qu.: 8534.0 3rd Qu.: 8082.48
## Max. :2020-11-20 Max. :19650.0 Max. :19892.0 Max. :19010.00
## Close Volume.BTC Volume.USD
## Min. : 120.0 Min. : 0 Min. :0.000e+00
## 1st Qu.: 470.5 1st Qu.: 5437 1st Qu.:2.873e+06
## Median : 3888.0 Median : 8285 Median :3.278e+07
## Mean : 4778.2 Mean : 11120 Mean :6.812e+07
## 3rd Qu.: 8299.5 3rd Qu.: 13771 3rd Qu.:9.201e+07
## Max. :19650.0 Max. :160541 Max. :1.238e+092014
btc %>%
filter(Date >= "2014-12-01") %>%
filter(Date <= "2014-12-31") %>%
summary(btc)## Date Open High Low
## Min. :2014-12-01 Min. :300.0 Min. :340.0 Min. :300.0
## 1st Qu.:2014-12-08 1st Qu.:340.0 1st Qu.:340.0 1st Qu.:340.0
## Median :2014-12-16 Median :350.0 Median :350.0 Median :350.0
## Mean :2014-12-16 Mean :352.5 Mean :356.4 Mean :351.2
## 3rd Qu.:2014-12-23 3rd Qu.:372.5 3rd Qu.:377.6 3rd Qu.:365.2
## Max. :2014-12-31 Max. :378.0 Max. :398.0 Max. :378.0
## Close Volume.BTC Volume.USD
## Min. :340.0 Min. : 0.0000 Min. : 0.00
## 1st Qu.:340.0 1st Qu.: 0.0000 1st Qu.: 0.00
## Median :350.0 Median : 0.0000 Median : 0.00
## Mean :353.8 Mean : 0.6788 Mean : 251.83
## 3rd Qu.:372.5 3rd Qu.: 0.0125 3rd Qu.: 4.72
## Max. :378.0 Max. :15.0100 Max. :5675.072015
btc %>%
filter(Date >= "2015-01-01") %>%
filter(Date <= "2015-12-31") %>%
summary(btc)## Date Open High Low
## Min. :2015-01-01 Min. :120.0 Min. :184.0 Min. :109.9
## 1st Qu.:2015-04-02 1st Qu.:235.5 1st Qu.:238.3 1st Qu.:231.8
## Median :2015-07-02 Median :249.8 Median :255.8 Median :244.9
## Mean :2015-07-02 Mean :273.8 Mean :280.1 Mean :267.3
## 3rd Qu.:2015-10-01 3rd Qu.:289.0 3rd Qu.:294.1 3rd Qu.:284.5
## Max. :2015-12-31 Max. :463.6 Max. :500.0 Max. :453.3
## Close Volume.BTC Volume.USD
## Min. :120.0 Min. : 0 Min. : 0
## 1st Qu.:235.5 1st Qu.: 5148 1st Qu.: 1308791
## Median :249.8 Median : 6706 Median : 1690436
## Mean :274.0 Mean : 7968 Mean : 2340055
## 3rd Qu.:289.0 3rd Qu.: 8919 3rd Qu.: 2436511
## Max. :463.6 Max. :160541 Max. :712468372016
btc %>%
filter(Date >= "2016-01-01") %>%
filter(Date <= "2016-12-31") %>%
summary(btc)## Date Open High Low
## Min. :2016-01-01 Min. :367.0 Min. :373.0 Min. :350.9
## 1st Qu.:2016-04-01 1st Qu.:432.6 1st Qu.:435.8 1st Qu.:427.6
## Median :2016-07-01 Median :583.2 Median :590.5 Median :574.6
## Mean :2016-07-01 Mean :567.1 Mean :575.1 Mean :557.8
## 3rd Qu.:2016-09-30 3rd Qu.:667.4 3rd Qu.:675.4 3rd Qu.:651.0
## Max. :2016-12-31 Max. :981.9 Max. :988.9 Max. :950.5
## Close Volume.BTC Volume.USD
## Min. :367.0 Min. : 0 Min. : 0
## 1st Qu.:432.7 1st Qu.: 3973 1st Qu.: 2268601
## Median :584.1 Median : 5108 Median : 2734549
## Mean :568.5 Mean : 6012 Mean : 3356748
## 3rd Qu.:667.9 3rd Qu.: 6885 3rd Qu.: 3533321
## Max. :981.9 Max. :27320 Max. :170390922017
btc %>%
filter(Date >= "2017-01-01") %>%
filter(Date <= "2017-12-31") %>%
summary(btc)## Date Open High Low
## Min. :2017-01-01 Min. : 801.5 Min. : 833 Min. : 0.06
## 1st Qu.:2017-04-02 1st Qu.: 1195.8 1st Qu.: 1221 1st Qu.: 1171.70
## Median :2017-07-02 Median : 2564.0 Median : 2670 Median : 2472.00
## Mean :2017-07-02 Mean : 3969.6 Mean : 4153 Mean : 3777.41
## 3rd Qu.:2017-10-01 3rd Qu.: 4582.0 3rd Qu.: 4657 3rd Qu.: 4420.01
## Max. :2017-12-31 Max. :19650.0 Max. :19892 Max. :19010.00
## Close Volume.BTC Volume.USD
## Min. : 801.5 Min. : 1314 Min. :1.332e+06
## 1st Qu.: 1197.3 1st Qu.: 7783 1st Qu.:1.113e+07
## Median : 2574.8 Median :12218 Median :3.357e+07
## Mean : 4004.9 Mean :14866 Mean :8.326e+07
## 3rd Qu.: 4599.0 3rd Qu.:18952 3rd Qu.:7.349e+07
## Max. :19650.0 Max. :90954 Max. :1.238e+092018
btc %>%
filter(Date >= "2018-01-01") %>%
filter(Date <= "2018-12-31") %>%
summary(btc)## Date Open High Low
## Min. :2018-01-01 Min. : 3183 Min. : 3229 Min. : 3130
## 1st Qu.:2018-04-02 1st Qu.: 6350 1st Qu.: 6450 1st Qu.: 6252
## Median :2018-07-02 Median : 6905 Median : 7100 Median : 6668
## Mean :2018-07-02 Mean : 7548 Mean : 7759 Mean : 7263
## 3rd Qu.:2018-10-01 3rd Qu.: 8670 3rd Qu.: 8933 3rd Qu.: 8280
## Max. :2018-12-31 Max. :17099 Max. :17178 Max. :16251
## Close Volume.BTC Volume.USD
## Min. : 3183 Min. : 0 Min. : 0
## 1st Qu.: 6350 1st Qu.: 6092 1st Qu.: 39534415
## Median : 6894 Median :10000 Median : 68458545
## Mean : 7519 Mean :12659 Mean : 98755346
## 3rd Qu.: 8589 3rd Qu.:16138 3rd Qu.:128945499
## Max. :17099 Max. :85899 Max. :6582766092019
btc %>%
filter(Date >= "2019-01-01") %>%
filter(Date <= "2019-12-31") %>%
summary(btc)## Date Open High Low
## Min. :2019-01-01 Min. : 3359 Min. : 3383 Min. : 3338
## 1st Qu.:2019-04-11 1st Qu.: 5078 1st Qu.: 5221 1st Qu.: 5014
## Median :2019-07-08 Median : 7952 Median : 8118 Median : 7621
## Mean :2019-07-05 Mean : 7469 Mean : 7673 Mean : 7252
## 3rd Qu.:2019-10-04 3rd Qu.: 9589 3rd Qu.: 9919 3rd Qu.: 9350
## Max. :2019-12-31 Max. :12927 Max. :13868 Max. :12122
## Close Volume.BTC Volume.USD
## Min. : 3359 Min. : 361.7 Min. : 1438461
## 1st Qu.: 5135 1st Qu.: 6547.9 1st Qu.: 37406908
## Median : 7952 Median : 8951.9 Median : 65132772
## Mean : 7479 Mean :12267.1 Mean :100360898
## 3rd Qu.: 9589 3rd Qu.:14751.6 3rd Qu.:123476350
## Max. :12927 Max. :70145.6 Max. :8878874932020
btc %>%
filter(Date >= "2020-01-01") %>%
filter(Date <= "2020-11-20") %>%
summary(btc)## Date Open High Low
## Min. :2020-01-01 Min. : 4857 Min. : 5350 Min. : 3858
## 1st Qu.:2020-03-22 1st Qu.: 8730 1st Qu.: 8973 1st Qu.: 8521
## Median :2020-06-11 Median : 9522 Median : 9686 Median : 9281
## Mean :2020-06-11 Mean : 9823 Mean :10052 Mean : 9590
## 3rd Qu.:2020-08-31 3rd Qu.:10944 3rd Qu.:11180 3rd Qu.:10736
## Max. :2020-11-20 Max. :17822 Max. :18488 Max. :17765
## Close Volume.BTC Volume.USD
## Min. : 4857 Min. : 2305 Min. : 21216023
## 1st Qu.: 8758 1st Qu.: 7923 1st Qu.: 76254157
## Median : 9526 Median : 11301 Median :106099896
## Mean : 9856 Mean : 14291 Mean :135004002
## 3rd Qu.:10960 3rd Qu.: 16634 3rd Qu.:163703447
## Max. :18143 Max. :117495 Max. :608984267Display Scatter Plots & Find Correlations
btc_cor <- btc[, 2:7]
chart.Correlation(btc_cor)
In the above plot:
- The distribution of each variable is shown on the diagonal.
- On the bottom of the diagonal : the bivariate scatter plots with a fitted line are displayed
- On the top of the diagonal : the value of the correlation plus the significance level as stars
- Each significance level is associated to a symbol : p-values(0, 0.001, 0.01, 0.05, 0.1, 1) <=> symbols(“”, “”, “”, “.”, " “)
These graphs show us that there is very little difference between the ‘Open’, ‘High’, ‘Low’, and ‘Close’ figures, so we are only going to remove all but the ‘Close’ variable and retest.
# Remove the Open, High and Low variables
btc <- subset(btc, select = -c(Open, High, Low))
# Rename Close variable to Price
names(btc)[names(btc) == "Close"] <- "Price"
# Check
head(btc)## Date Price Volume.BTC Volume.USD
## 2170 2014-12-01 370.0 0.05656 19.53
## 2169 2014-12-02 378.0 15.01000 5675.07
## 2168 2014-12-03 378.0 0.54660 206.52
## 2167 2014-12-04 377.1 0.01000 3.77
## 2166 2014-12-05 377.1 0.00000 0.00
## 2165 2014-12-06 378.0 0.01500 5.67# Re-run correlation matrix &^ scatter plots
btc_cor <- btc[, 2:4]
chart.Correlation(btc_cor)
- The strongest linear relationship is between Volume.BTC and Volume.USD—as shown by the straight diagonal line and correlation of 0.77. - The correlation between Price and Volume.USD is also high at 0.7 but the correlation betweeen Price and Volume.BTC is just inside the ideal threshold of < 0.4 or > -0.4 at 0.35. Since Volume.BTC it is the only x-variable that has a correlation inside the ideal threshold and it’s with the y-variable we’re trying to predict—Bitcoin-further evaluation will be needed with additional data sets to run a linear regression and more advanced data science methods for predicting the price of Bitcoin. - All values have a p-value of ~0 (as represented by the ***) which means they are have a significant difference which isn’t caused by chance
Bitcoin to Google Search Interest
Add Google Search Data
# Sourced from https://trends.google.com/trends/explore?date=2014-12-01%202020-11-20&geo=US&q=Bitcoin
# Read CSV data into R
search <- read.csv("btc_search_volume.csv", header = TRUE)
# Convert Date factor into date format
search$Date <- as.Date(search$Date, format = "%Y-%m-%d")
# Display the first 6 elements to ensure that the data is read properly
head(search)## Date Volume
## 1 2014-12-01 3
## 2 2015-01-01 3
## 3 2015-02-01 3
## 4 2015-03-01 2
## 5 2015-04-01 2
## 6 2015-05-01 2Display Google Search Volume Summary
summary(search)## Date Volume
## Min. :2014-12-01 Min. : 2.00
## 1st Qu.:2016-05-24 1st Qu.: 3.00
## Median :2017-11-16 Median : 8.00
## Mean :2017-11-15 Mean : 10.08
## 3rd Qu.:2019-05-08 3rd Qu.: 11.00
## Max. :2020-11-01 Max. :100.00Format BTC Database
# Create new database that filters for the first day of each month (since this is the only search volume data only available)
# Create startpoints function
startpoints <- function (x, on = "months", k = 1) {
head(endpoints(x, on, k) + 1, -1)
}
btc_m <- btc
btc_m <- btc_m[startpoints(btc_m$Date, on = "months"),]
head(btc_m)## Date Price Volume.BTC Volume.USD
## 2170 2014-12-01 370.00 0.05656 19.53
## 2139 2015-01-01 340.00 0.00000 0.00
## 2108 2015-02-01 227.00 7401.43000 1640635.74
## 2080 2015-03-01 258.78 4967.35000 1255486.27
## 2049 2015-04-01 247.25 9529.31000 2333725.21
## 2019 2015-05-01 234.01 9697.40000 2287292.43Combine BTC Month and Google Search Data
btc_search <- data.frame(btc_m, Volume.Search = search$Volume)
# Check
head(btc_search)## Date Price Volume.BTC Volume.USD Volume.Search
## 2170 2014-12-01 370.00 0.05656 19.53 3
## 2139 2015-01-01 340.00 0.00000 0.00 3
## 2108 2015-02-01 227.00 7401.43000 1640635.74 3
## 2080 2015-03-01 258.78 4967.35000 1255486.27 2
## 2049 2015-04-01 247.25 9529.31000 2333725.21 2
## 2019 2015-05-01 234.01 9697.40000 2287292.43 2Create Dual Y-Axis Comparable Graph
ds <- data.frame(Date = btc_search$Date, btc_search$Price, btc_search$Volume.Search)
ay <- list(
tickfont = list(color = "Orange"),
overlaying = "y",
side = "right",
title = "Weighted Search Volume"
)
fig <- plot_ly(ds, x = ~Date)
fig <- fig %>% add_lines(x = ~Date, y = ~btc_search$Price, name = "Bitcoin")
fig <- fig %>% add_lines(y = ~btc_search$Volume.Search, name = "Google Search", yaxis = "y2")
fig <- fig %>% layout(
title = "Bitcoin Price vs. Google Search Volume",
yaxis2 = ay,
xaxis = list(
rangeselector = list(
buttons = list(
list(
count = 3,
label = "3 mo",
step = "month",
stepmode = "backward"),
list(
count = 6,
label = "6 mo",
step = "month",
stepmode = "backward"),
list(
count = 1,
label = "1 yr",
step = "year",
stepmode = "backward"),
list(
count = 1,
label = "YTD",
step = "year",
stepmode = "todate"),
list(step = "all"))),
rangeslider = list(type = "date")),
yaxis = list(title = "Price (USD)")
)
figbtc_search_cor <- btc_search[, 2:5]
chart.Correlation(btc_search_cor)
All correaltions are above the 0.4 threshold so I am going to remove the highest (Volume.USD) and run again.
btc_search_cor <- btc_search[, c(2,3,5)]
chart.Correlation(btc_search_cor)
The Volume.Search variable is still above the 0.4 threshold which’d bring us back to the original correlation matrix above (btc variables only). We can infer further evaluation is needed to predict the price of Bitcoin.
Compare Cryptocurrencies
For simplicity, I am representing the price with the ‘Close’ variable and eliminating ‘Open’, ‘High’, and ‘Low’, for all cryptocurrencies and future finanical x.
Format Coin Data
Bitcion Data
# Rename Close variable to Price
names(btc)[names(btc) == "Price"] <- "BTC.Price"
# Rename Volume.USD to Currency
names(btc)[names(btc) == "Volume.USD"] <- "BTC.Volume.USD"
head(btc)## Date BTC.Price Volume.BTC BTC.Volume.USD
## 2170 2014-12-01 370.0 0.05656 19.53
## 2169 2014-12-02 378.0 15.01000 5675.07
## 2168 2014-12-03 378.0 0.54660 206.52
## 2167 2014-12-04 377.1 0.01000 3.77
## 2166 2014-12-05 377.1 0.00000 0.00
## 2165 2014-12-06 378.0 0.01500 5.67Ethereum Data
# Read CSV data into R
eth_data <- read.csv("Coinbase_ETHUSD_d.csv", header = TRUE)
# Order rows by date
eth = eth_data[order(eth_data$Date),]
# Remove Timestamp & Symbol
eth = subset(eth, select = -c(Unix.Timestamp, Symbol))
# Convert Date factor in date format
eth$Date <- as.Date(eth$Date, format = "%Y-%m-%d")
# Remove the Open, High and Low variables
eth <- subset(eth, select = -c(Open, High, Low))
# Rename Close variable to Price
names(eth)[names(eth) == "Close"] <- "ETH.Price"
# Rename Volume.USD to Currency
names(eth)[names(eth) == "Volume.USD"] <- "ETH.Volume.USD"
# Display the first 6 elements to ensure that the data is read properly
head(eth)## Date ETH.Price Volume.ETH ETH.Volume.USD
## 1627 2016-05-27 11.25 13749.03 151147.98
## 1626 2016-05-28 11.93 15810.34 180822.02
## 1625 2016-05-29 12.34 3481.30 42228.37
## 1624 2016-05-30 12.41 4110.59 51655.95
## 1623 2016-05-31 14.00 5654.91 76994.75
## 1622 2016-06-01 13.93 10369.02 145746.12Litecoin Data
# Read CSV data into R
ltc_data <- read.csv("Coinbase_LTCUSD_d.csv", header = TRUE)
# Order rows by date
ltc = ltc_data[order(ltc_data$Date),]
# Remove Timestamp & Symbol
ltc = subset(ltc, select = -c(Unix.Timestamp, Symbol))
# Convert Date factor into date format
ltc$Date <- as.Date(ltc$Date, format = "%Y-%m-%d")
# Remove the Open, High and Low variables
ltc <- subset(ltc, select = -c(Open, High, Low))
# Rename Close variable to Price
names(ltc)[names(ltc) == "Close"] <- "LTC.Price"
# Rename Volume.USD to Currency
names(ltc)[names(ltc) == "Volume.USD"] <- "LTC.Volume.USD"
# Display the first 6 elements to ensure that the data is read properly
head(ltc)## Date LTC.Price Volume.LTC LTC.Volume.USD
## 1539 2016-08-23 3.95 443.86 1737.31
## 1538 2016-08-24 3.84 5083.21 19247.53
## 1537 2016-08-25 3.81 5085.64 19276.60
## 1536 2016-08-26 3.81 3336.89 12746.27
## 1535 2016-08-27 3.78 1140.72 4295.72
## 1534 2016-08-28 3.72 1904.98 7111.87Ripple Data
# Read CSV data into R
xrp_data <- read.csv("Bitstamp_XRPUSD_d.csv", header = TRUE)
# Order rows by date
xrp = xrp_data[order(xrp_data$Date),]
# Remove Timestamp & Symbol
xrp = subset(xrp, select = -c(Unix.Timestamp, Symbol))
# Convert Date factor into date format
xrp$Date <- as.Date(xrp$Date, format = "%Y-%m-%d")
# Remove the Open, High and Low variables
xrp <- subset(xrp, select = -c(Open, High, Low))
# Rename Close variable to Price
names(xrp)[names(xrp) == "Close"] <- "XRP.Price"
# Rename Volume.USD to Currency
names(xrp)[names(xrp) == "Volume.USD"] <- "XRP.Volume.USD"
# Display the first 6 elements to ensure that the data is read properly
head(xrp)## Date XRP.Price Volume.XRP XRP.Volume.USD
## 1392 2017-01-17 0.00683 4509598.6 30673.69
## 1391 2017-01-18 0.00680 5535202.5 38018.93
## 1390 2017-01-19 0.00684 2951225.7 19882.33
## 1389 2017-01-20 0.00660 1713629.9 11374.15
## 1388 2017-01-21 0.00684 2086817.3 13955.92
## 1387 2017-01-22 0.00678 573857.3 3878.76Monero Data
# Read CSV data into R
xmr_data <- read.csv("Poloniex_XMRUSD_d.csv", header = TRUE)
# Order rows by date
xmr = xmr_data[order(xmr_data$Date),]
# Remove Timestamp & Symbol
xmr = subset(xmr, select = -c(Unix.Timestamp, Symbol))
# Convert Date factor into date format
xmr$Date <- as.Date(xmr$Date, format = "%Y-%m-%d")
# Remove the Open, High and Low variables
xmr <- subset(xmr, select = -c(Open, High, Low))
# Rename Close variable to Price
names(xmr)[names(xmr) == "Close"] <- "XMR.Price"
# Rename Volume.USD to Currency
names(xmr)[names(xmr) == "Volume.USD"] <- "XMR.Volume.USD"
# Display the first 6 elements to ensure that the data is read properly
head(xmr)## Date XMR.Price Volume.XMR XMR.Volume.USD
## 2110 2015-01-29 0.3107 68.21 23.86
## 2109 2015-01-30 0.3027 230.00 80.20
## 2108 2015-01-31 0.2896 82.10 24.55
## 2107 2015-02-01 0.2896 0.00 0.00
## 2106 2015-02-02 0.3029 59.58 17.89
## 2105 2015-02-03 0.3311 254.39 90.77Dash Data
# Read CSV data into R
dash_data <- read.csv("Poloniex_DASHUSD_d.csv", header = TRUE)
# Order rows by date
dash = dash_data[order(dash_data$Date),]
# Remove Timestamp & Symbol
dash = subset(dash, select = -c(Unix.Timestamp, Symbol))
# Convert Date factor into date format
dash$Date <- as.Date(dash$Date, format = "%Y-%m-%d")
# Remove the Open, High and Low variables
dash <- subset(dash, select = -c(Open, High, Low))
# Rename Close variable to Price
names(dash)[names(dash) == "Close"] <- "DASH.Price"
# Rename improperly formatted Volume.DAS & Volume.HUSD
names(dash)[names(dash) == "Volume.DAS"] <- "Volume.DASH"
names(dash)[names(dash) == "Volume.HUSD"] <- "DASH.Volume.USD"
# Display the first 6 elements to ensure that the data is read properly
head(dash)## Date DASH.Price Volume.DASH DASH.Volume.USD
## 2091 2015-02-17 10 0.001992 0.01992
## 2090 2015-02-18 10 0.000000 0.00000
## 2089 2015-02-19 10 0.000000 0.00000
## 2088 2015-02-20 10 0.000000 0.00000
## 2087 2015-02-21 10 0.000000 0.00000
## 2086 2015-02-22 10 0.000000 0.00000Filter Rows for Consistency
# Total number of rows (from XRP database since it has the fewest historical data points)
c_rows <- 1392
# Format Bitcoin
btc_c <- tail(btc,n=c_rows)
head(btc_c)## Date BTC.Price Volume.BTC BTC.Volume.USD
## 1392 2017-01-17 907.58 7258.23 6401902
## 1391 2017-01-18 880.00 6654.70 5878776
## 1390 2017-01-19 901.00 5127.03 4602695
## 1389 2017-01-20 894.05 3073.67 2761267
## 1388 2017-01-21 925.36 4545.67 4173036
## 1387 2017-01-22 924.98 4915.78 4556300# Ethereum
eth <- tail(eth,n=c_rows)
head(eth)## Date ETH.Price Volume.ETH ETH.Volume.USD
## 1392 2017-01-17 10.20 69235.88 705202.8
## 1391 2017-01-18 10.23 62085.98 635612.0
## 1390 2017-01-19 10.47 46097.26 477203.6
## 1389 2017-01-20 10.65 60007.76 637465.4
## 1388 2017-01-21 10.99 58646.70 635251.4
## 1387 2017-01-22 10.78 48319.33 525789.5# Litecoin
ltc <- tail(ltc,n=c_rows)
head(ltc)## Date LTC.Price Volume.LTC LTC.Volume.USD
## 1392 2017-01-17 3.93 1774.48 7050.37
## 1391 2017-01-18 3.89 2195.30 8536.72
## 1390 2017-01-19 3.88 3095.56 12060.09
## 1389 2017-01-20 3.88 893.27 3490.55
## 1388 2017-01-21 3.92 620.48 2425.73
## 1387 2017-01-22 3.86 1711.05 6664.88# Ripple
xrp <- tail(xrp,n=c_rows)
head(xrp)## Date XRP.Price Volume.XRP XRP.Volume.USD
## 1392 2017-01-17 0.00683 4509598.6 30673.69
## 1391 2017-01-18 0.00680 5535202.5 38018.93
## 1390 2017-01-19 0.00684 2951225.7 19882.33
## 1389 2017-01-20 0.00660 1713629.9 11374.15
## 1388 2017-01-21 0.00684 2086817.3 13955.92
## 1387 2017-01-22 0.00678 573857.3 3878.76# Monero
xmr <- tail(xmr,n=c_rows)
head(xmr)## Date XMR.Price Volume.XMR XMR.Volume.USD
## 1392 2017-01-16 10.51 11549.89 124038.82
## 1391 2017-01-17 12.33 14986.29 172577.12
## 1390 2017-01-18 12.00 15813.59 192223.22
## 1389 2017-01-19 12.15 6526.12 78925.65
## 1388 2017-01-20 11.88 5532.43 66342.50
## 1387 2017-01-21 12.09 6985.26 84206.72# Dash
dash <- tail(dash,n=c_rows)
head(dash)## Date DASH.Price Volume.DASH DASH.Volume.USD
## 1392 2017-01-16 12.82 417.30 5253.05
## 1391 2017-01-17 14.33 2220.10 30577.09
## 1390 2017-01-18 14.00 2455.66 34117.04
## 1389 2017-01-19 15.16 2766.26 40779.23
## 1388 2017-01-20 15.25 2852.34 43349.21
## 1387 2017-01-21 14.95 721.51 10816.23# Check row count (find formula that'll return the number of rows for each)Normalize the Data
Compare With Traditional Investment Vehichles (Equities & bonds)
(representde by the Dow jones and US treasury bonds)
Evaluate Bitcoin Price Compared to the Top 5 Currencies
#plot_top_5_currencies()Display Large Scatter Plots
# Map the correlation between Bitcoin's closing price and volume traded (in USD)
#plot(btc$Price, btc$Volume.USD, main = 'Price to Volume (USD) Comparison', xlab = 'Closing Price', ylab = 'Volume (USD)')# Map the correlation between Bitcoin's closing price and volume traded (in BTC)
#plot(btc$Price, btc$Volume.BTC, main = 'Price to Volume (BTC) Comparison', xlab = 'Closing Price', ylab = 'Volume (BTC)')